INDEX

Explanations

questions or rhetorical queries

oai_token-act-pair · gpt-3.5-turbo

questions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yak

-0.66

ishable

-0.65

 striped

-0.64

apan

-0.64

lock

-0.63

 soap

-0.63

 foam

-0.62

 worm

-0.62

apter

-0.62

 docks

-0.62

POSITIVE LOGITS

 Surely

1.03

 Wouldn

1.02

 Nope

0.99

 Certainly

0.96

ï¿½ï¿½ï¿½ï¿½

0.94

 Conversely

0.92

 Presumably

0.92

 Perhaps

0.91

 Probably

0.90

Why

0.89

Activations Density 0.110%

questions or rhetorical queries

questions

No Comments

No Known Activations